基于灰度方差的指纹图像分割方法python_指纹分割 python

import copy
import math

import cv2
import numpy as np

# 基于灰度方差的指纹图像分割方法,8×8小块
img = cv2.imread('021_9_2.png')  # 读取图片
img_0 = img[:, :, 0]  # 取图像中的一个通道
W = 8  # 小块的尺寸
row = img_0.shape[0]
col = img_0.shape[1]
num_row_W = math.ceil(row / W)
num_col_W = math.ceil(col / W)
Var = np.empty((num_row_W, num_col_W))  # 方差矩阵
for i in range(0, row, W):
    for j in range(0, col, W):
        row_var = int(i / W)
        col_var = int(j / W)
        r_row = min(i + W, row)
        r_col = min(j + W, col)
        tmp = img_0[i:r_row, j:r_col]
        M = np.sum(tmp) / (r_row - i) / (r_col - j)  # 计算小块平均值
        tmp = tmp - M
        Var[row_var, col_var] = np.sum(tmp * tmp) / (r_row - i) / (r_col - j)  # 计算方差

threshold = 100  # 阈值为100,根据不同指纹图像设置
img_cut = copy.deepcopy(img)
# 将方差小于阈值的方块置为0
for i in range(0, row, W):
    for j in range(0, col, W):
        row_var = int(i / W)
        col_var = int(j / W)
        if Var[row_var, col_var] < threshold:
            img_cut[i:min(i + W, row), j:min(j + W, col), :] = 0

# 寻找指纹的矩形边界
min_row = 10000
max_row = -1
min_col = 10000
max_col = -1
for i in range(0, row, W):
    for j in range(0, col, W):
        if img_cut[i, j, 0] != 0:
            min_row = min(i, min_row)
            max_row = max(i, max_row)
            min_col = min(j, min_col)
            max_col = max(j, max_col)
im = img[min_row:max_row, min_col:max_col, :]  # 原指纹图像切割
im_cut = img_cut[min_row:max_row, min_col:max_col, :]  # 指纹图像空白区域置0切割

cv2.imshow('im', im)
cv2.waitKey(0)
cv2.destroyAllWindows()

cv2.imshow('im_cut', im_cut)
cv2.waitKey(0)
cv2.destroyAllWindows()
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58

原指纹图像：

原指纹图像切割：

指纹图像空白区域置0切割：